Catena 122 (2014) 180–195

Contents lists available at ScienceDirect

Catena

journal homepage: www.elsevier.com/locate/catena

Short-term geomorphic analysis in a disturbed fluvial environment by fusion of LiDAR, colour bathymetry and dGPS surveys

J. Moretto a,⁎,E.Rigona,L.Maob,F.Delaia,L.Piccoa, M.A. Lenzi a a Department of Land, Environment, Agriculture and Forestry, University of Padova, Padova, Italy b Department of Ecosystems and Environment, Pontificia Universidad Catolica de Chile, Santiago, Chile article info abstract

Article history: Objective: Estimating river's underwater bed elevations is a necessary but challenging task. The objective of this Received 11 November 2013 study is to develop a revised approach to generate accurate and detailed Digital Terrain Models (DTMs) of a river Received in revised form 27 June 2014 reach by merging LiDAR data for the dry area, with water depth indirectly derived from aerial imagery for wet Accepted 30 June 2014 areas. Available online xxxx Methods: This approach was applied along three sub-reaches of the Brenta River (Italy) before and after two major flood events. A regression model relating water depth and intensity of the three colour bands derived Keywords: from aerial photos, was implemented. More than 2400 in-channel depth calibration points were taken using a Fluvial processes Gravel-bed river differential Global Positioning System (dGPS) along a wide range of underwater bed forms. LiDAR data Results: The resulting DTMs closely matched the field-surveyed bed surface, and allowed to assess that a 10-year Colour bathymetry recurrence interval flood generated a predominance of erosion processes. Erosion dominated in the upper part of Floods the study segment (−104,082 m3), whereas a near-equilibrium is featured on the lower reach (−45,232 m3). DoD The DTMs allowed the detection of processes such as riffle–pool downstream migration, and the progressive scour of a pool located near a rip-rap. Conclusion: The presented approach provides an adequate topographical description of the river bed to explore channel adjustments due to flood events. Practice: Combining colour bathymetry and dGPS surveys proved to represent a useful tool for many fluvial en- gineering, ecology, and management purposes. Implications: The proposed approach represents a valuable tool for river topography description, river manage- ment, ecology and restoration purposes, when bathymetric data are not available. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Three-dimensional and high-resolution representations of river bed morphology are used in many applications such as hydraulic and cellu- The study of river morphology and dynamics is essential for under- lar modelling (e.g. Rumsby et al., 2008), evaluation of climate change standing the factors determining sediment erosion, transport and depo- impacts on river systems (e.g. Rumsby and Macklin, 1994), flood hazard sition processes. Natural (e.g. climatic and hydrological variations) and management (Fewtrell et al., 2011; Macklin and Rumsby, 2007; anthropic factors (e.g. water captures, grade-control works, gravel min- Sampson et al., 2012), assessment of erosion and deposition areas ing, deforestation) can act at both the reach- and basin-scales changing along the river corridor (Lane et al., 2007; Picco et al., 2013; Stover the magnitude and timing of these processes (Buffington, 2012). Geo- and Montgomery, 2001). Calculating sediment budgets, estimating morphic variations at the reach scale are a direct consequence of sedi- transport rates, and understanding changes in sediment storage are ment erosion and deposition processes, which are in turn influenced also fundamental aspects to quantify geomorphological changes due by the size and volume of sediment supply, transport capacity of the to flood events and changes in flow regime (Ashmore and Church, flow, and local topographic constraints. The actual ability to quantify 1998; Wheaton et al., 2013). the interaction of these processes is limited by the difficulty of collecting The traditional techniques of terrain survey (e.g. total station de- high spatial resolution data in river environments. Traditional ap- vices, differential Global Positioning System - dGPS; Brasington et al., proaches, based on the application of hydraulic formulas at cross- 2000) in the evaluation of morphological changes across large areas sections, fail when aimed at describing non-uniform natural conditions. have so far demonstrated to be expensive, time-consuming and difficult to apply in zones with limited accessibility. Some innovative methods are good alternatives for producing high-resolution Digital Terrain ⁎ Corresponding author at: Agripolis Campus, Viale dell'Università, 16. 35020 - Legnaro Padova, Italy. Fax: +39 0 498272750. Models (DTMs) of fluvial systems. Recent studies on morphological E-mail address: [email protected] (J. Moretto). channel changes have used passive remote sensing techniques such as

http://dx.doi.org/10.1016/j.catena.2014.06.023 0341-8162/© 2014 Elsevier B.V. All rights reserved. J. Moretto et al. / Catena 122 (2014) 180–195 181 digital image processing (e.g. Forward Image Model, Legleiter and changes occurring in river systems over time (e.g. Lane et al., 1994). Roberts, 2009), digital photogrammetry (Brasington et al., 2003; An important component to be evaluated in DEMs is uncertainty, Dixon et al., 1998; Heritage et al., 1998; Lane et al., 2010), active sensors which can be influenced by many factors. The most decisive sources of including Laser Imaging Detection and Ranging (LiDAR) (e.g. Brasington error include survey point quality, sampling strategy, surface topo- et al., 2012; Hicks, 2012; Hicks et al., 2002, 2006; Kinzel et al., 2007), and graphic complexity and interpolation methods (Milan et al., 2011; acoustic methods (e.g. Muste et al., 2012; Rennie, 2012). Panissod et al., 2009; Wheaton et al., 2010). Total uncertainty is usually The main difficulty related to the production of precise DTMs with no- derived from the classical statistical theory of errors (Taylor, 1997) bathymetric sensors concerns the absorption of natural (solar) or artificial where an estimation of DEM accuracy based on survey data is used as (LiDAR) electromagnetic radiation in the wetted portion of the river chan- a surrogate for DEM quality (Milan et al., 2007). nels. The capacity of the electromagnetic signal to pass through water, be This paper implements a revised approach to generate accurate and reflected from the bed and reach a sensor depends on water surface tex- detailed Digital Terrain Models (DTMs) of a river reach by merging ture (pleating, reflexes, etc.), water column (depth and turbidity), and na- LiDAR data for dry areas, with water depth estimates for wet areas. ture of channel bed (substrate type and presence of algae; Marcus, 2012; The main objective consists in the evaluation of morphological patterns Marcus and Fonstad, 2008). Even if different electro-magnetic (EM) of change in three sub-reaches of the Brenta River as a consequence of wavelengths are absorbed by water at different degrees (Smith and two consecutive major floods occurred in November and December Vericat, 2013; Smith et al., 2012), LiDAR signal proved to be adequately 2010. Specific aims related with the proposed approach and targeted reliable to assess channel topography “where flow is sufficiently shallow to achieve an adequate topographic description of the river bed are: to that water depth does not distort the laser” (Charlton et al., 2003), as late- determine physical and empirical relations between local channel ly confirmed also by Cavalli and Tarolli (2011). depths and photo colour intensity; to identify and to filter the factors Only a few tools have proved able to provide an accurate and high- which increase uncertainty in the final DTM, in order to obtain Hybrid resolution measure of submerged bed surface. Moreover, the precision DTMs (HDTMs) at high resolution and low uncertainty. of the surveyed data decreases as water depth increases. Bathymetric LiDAR sensors have recently been developed and should enable the sur- 2. Study area vey of underwater bed surfaces. Nevertheless, they feature high costs, relatively low resolutions, and data quality comparable to photogram- The Brenta River is located in the South-Eastern Italian Alps, has a metric techniques (Hilldale and Raff, 2008). Progress in LiDAR acquisi- drainage basin of approximately 1567 km2 and a length of 174 km. tion of topographic information from submerged areas has been The average annual precipitation, mainly concentrated in spring and au- achieved with a new technology called Experimental Advanced Air- tumn, is about 1100 mm. The geology of the area is rather complex and borne Research LiDAR system (EAARL), which records the full wave- includes limestone, dolomite, gneiss, phyllite, granite and volcanic form of returning laser pulse. Even if this system is affected by rocks. environmental conditions (e.g. turbulence in the pool, bubbles in the The study reach is 19.2 km-long and lies between Bassano Del Grap- water column, turbidity, and low-bottom albedo) and by post- pa and Piazzola sul Brenta (Fig. 1). The dominant morphologies are processing algorithms, its accuracy appears comparable to what is ob- wandering and braided, the active channel width varies between 300 m tained using airborne terrestrial near-infrared LiDARs (Kinzel et al., and 800 m, and the average slope is about 0.0036 m/m. Within this 2013; McKean et al., 2009). study reach, three sub-reaches 1.5 km-long and 5 km apart were se- Surveys of wet areas can thus be approached using two photogram- lected as representative of the upper- middle- and down-stream part metric techniques (manual or automatic) which are able to produce a of the study area and named according to the nearby villages: Nove, cloud of elevation points (Fryer, 1983; Rinner, 1969), or with a tech- Friola and Fontaniva (Fig. 1). The upstream sub-reach (Nove) has a nique based on the calibration of a depth–reflectance relationship of im- single straightened channel morphology with an average width of ages, which can be in greyscale (e.g. Winterbottom and Gilvear, 1997), around 300 m. By contrast, Friola shows a more complex morpholog- coloured (e.g. Carbonneau et al., 2006; Moretto et al., 2013a; Williams ical pattern, with the braided channel accounting for high levels of et al., 2011, 2013, 2014) or multispectral (Legleiter, 2011; Marcus vegetation density and an average width of 500 m. In the down- et al., 2003). Both solutions need a field survey, contemporary to the stream sub-reach, called Fontaniva, the river is 800 m wide, braided flight, to provide calibration depth points. and features many fluvial islands. The depth–refl ectance relationship can be defined using an empiri- The Brenta river basin has suffered centuries of disturbances, mostly cal equation, using one or more bands (e.g. Legleiter et al., 2009), or ac- due to deforestation and reforestation phases. The water course has cording to the Beer–Lambert law. For the latter case, the amount of light long been regulated for hydroelectric power generation and irrigation absorbed by a transparent material is considered to be proportional to purposes and dams were built in many parts of the drainage basin, the distance of the light travelling through that material (Carbonneau intercepting sediment from more than 40% of the drainage area. More- et al., 2006): over, between 1953 and 1985, gravel was intensively quarried in the main channel and, starting in the 1930s, effective erosion and torrent ¼ −cx ð Þ Iout Iine 1 control works were executed in the upper basin (Bathurst et al., 2003; Conesa-Garcìa and Lenzi, 2013; Lenzi, 2006; Lenzi et al., 2003; Rigon where Iin is the incoming intensity [no units], Iout is the outgoing intensi- et al., 2008, 2012; Surian et al., 2009). Human interventions, especially ty [no units], c is the rate of light absorption derived multiplying the during the second half of the 20th century, have considerably altered molar absorptivity [L mol−1 cm−1] by the solution concentration [mol the sediment budget of many Alpine rivers (Comiti, 2011; Comiti L−1],andx is the distance [cm]. et al., 2011; Mao and Lenzi, 2007; Mao et al., 2009; Picco et al., 2013). However, the calibration of a depth–reflectance relationship be- As a result of these impacts, the average riverbed width of the Brenta comes challenging when the channel bottom is composed by sediment has narrowed from 442 m at the beginning of the 1800s, to 196 m in of different sizes, periphyton, woody debris, vegetation, senescent veg- 2010, and channel incision has ranged from 2 to 8 m, especially due to etation, artificial artefacts, etc. In fact, the composition of the channel the effects of gravel quarrying which ended only during the 1990s bed can strongly affect the local reflectance of the wet areas (Legleiter (Kaless et al., 2014; Moretto et al., 2012a,b, 2013b; Surian and Cisotto, et al., 2009), introducing a greater variability which should be taken 2007). In recent times, a new adjustment phase seems to be taking into account in the depth-colour model. place (channel widened to 215 m in 2011) as evidenced by the Once reliable digital elevation models (DEMs) have been obtained, it expanding trend of the active channel with a contemporary increase is possible to detect and interpret, in a quantitative way, geomorphic in vegetated islands over the last twenty years (Moretto et al., 2012a, 182 J. Moretto et al. / Catena 122 (2014) 180–195

Fig. 1. General view of the Brenta river and the study sub-reaches: Nove, Friola and Fontaniva. b, 2013b). The recent evolutionary dynamics considerably differ from recovery, especially in the downstream sub-reach, Fontaniva. However, those observed in the past. Since the abandonment of gravel mining ac- this trend is still unstable and not distributed along the whole study tivities on the river bed (1990s), there has been a partial morphological reach. In the upstream area, incision processes and a widening trend

Fig. 2. Hydrograph of the study period (average daily discharges as measured at the Barzizza gauging station). Recurrence interval of the two highest flood peaks has been reached 8 years (Novembre 2010) and 10 years (Decembre 2010) respectively. J. Moretto et al. / Catena 122 (2014) 180–195 183 of the active channel as a result of bank erosion are still present for each sub-reach) were produced for each year, and the volumetric (Moretto et al., 2012a,b, 2013b). surface changes and relative uncertainty calculated (see next sections). Two flood events have occurred between the LiDAR flights, conduct- ed in August 2010 and on April 2011 (Fig. 2). The November 2010 3.1. LiDAR data and field surveys flood reached a maximum average daily discharge of 720 m3/s, with a slower drop in the water level compared to the second flood event Two LiDAR surveys were conducted on the 23rd of August 2010 by occurred in December 2010, which reached the highest discharge Blom GCR Spa with an OPTECH ALTM Gemini sensor, and on the 24th of the last 10 years (maximum average daily discharge of 759 m3/s). of April 2011 by OGS Company with a RIEGL LMS-Q560 sensor (flying The recurrence intervals of these floods were estimated from the max- height ~ 850 m). For each LiDAR survey, a point density able to generate imum annual values of the mean daily water discharge over 79 hydro- digital terrain models with 0.50 m resolution was commissioned. The logical years. Among various tested probability distributions, the average vertical error of the LiDAR was evaluated through dGPS points Gumbel distribution (OLS) demonstrated the best performance comparison on the final elevation model. The LiDAR data were taken (Kolmogoroff test). Taking into account the Gumbel distribution and along with a series of RGB aerial photos with 0.15 m of pixel resolution. 90% confidence limits, it was possible to establish the flood values asso- The surveys were conducted with clear weather conditions and low hy- ciated with the probability of occurrence (Kaless et al., 2011, 2014; Lenzi draulic channel levels. An in-channel dGPS survey was performed, tak- et al., 2010). ing different depth levels in a wide range of morphological units. A The first flood was caused by prolonged and heavy rainfall (300 mm) total of 882 points in 2010 and 1526 points in 2011 were surveyed. between the 31st of October and 2nd of November 2010 (Fig. 3), and Depth ranges of surveyed calibration points were between 0.20 m and featured a recurrence interval (RI) of about 8 years. The second flood, 1.60 m. It is important to note that the dGPS survey was performed si- originated by intensive precipitations between the 21st and 26th of De- multaneously to LiDAR data acquisition to avoid additional sources of cember, had a R.I. of about 10 years. Rainfall exceeded 150 mm with errors. local maximums of 300–400 mm and the river registered (at the Finally, two cross-sections for each sub-reach were surveyed Barzizza station) higher hydrometric levels than the first flood event, through dGPS (average vertical error ± 0.025 m), measuring each sig- probably due to the greater soil saturation at basin scale and, more par- nificant topographical change and, at least, one point per metre length. ticularly, to the fact that a major reservoir (Corlo) had already been filled by previous precipitations. 3.2. Dataset preparation

The raw LiDAR point clouds were analysed and the ground surface 3. Material and methods was identified through an automatic filtering algorithm (TerraScan, Microstation Application®). In critical areas, such as near bridges, man- In order to create an accurate digital terrain model accounting for re- ual checks were utilised. The aerial photos were georeferenced and liable river bed elevations, a regression model was calibrated between corrected by applying a brightness analysis in a semi-automatic ap- water depth and Red, Green and Blue (RGB) bands obtained from aerial proach: the tool “reference” of TerraPhoto (Microstation application®) images acquired during the LiDAR surveys. Water depth was calculated was used to combine aerial photos with contemporary LiDAR data and indirectly as the difference between water surface (estimated from the flight trajectories. The corrected photos were joined (ESRI® ArcMap interpolation of selected LiDAR points; see details in Section 3.2) and 10) and the pixel size was resampled from 0.15 m to 0.50 m to minimise channel bed elevation (measured with dGPS in the field). Hybrid Digital georeferencing errors and reduce possible strong colour variations due Terrain Models (HDTMs) were then created, by merging LiDAR to light reflection, exposed sediment, periphyton, shadows and (Section 3.4) points for dry areas and colour bathymetry-derived points suspended load. This represents a crucial point because poor photo for wet areas. Overall, three HDTMs were obtained for each year and georeferencing may significantly increase errors due to a wrong associ- each sub-reach (Nove, Friola and Fontaniva). ation between water depth and colour intensity. This computational process (Fig. 4) was divided into five principal Wet areas were digitised through a manual photo-interpretation steps: (A) LiDAR data and field survey, (B) dataset preparation, process. Along the edges of the digitised “wet areas”, LiDAR points (C) bathymetric model determination, (D) HDTMs creation and able to represent water surface elevation (Zw) were selected (to avoid (E) HDTMs validation. Finally, three DEMs of difference (DoDs — one points between wet and dry areas but above the water surface; e.g. on

Fig. 3. The Brenta river at Friola reach during November 2010 flood event. 184 J. Moretto et al. / Catena 122 (2014) 180–195

Fig. 4. HDTM creation process: (A) LiDAR data and field survey, (B) data preparation for process application, (C) bathymetric model determination, (D) hybrid DTM creation, (E) DTM validation. a vertical bank) on average every 10 m or in correspondence of signifi- best model (test points). Physical models based on Beer Lambert law cant slope changes. These points were then used to create a water sur- (Eq. (1)) were tested first. face elevation raster (i.e. Kriging interpolation). The error of the A ratio-based method was employed to detect changes in depth and resulting water surface was validated through 1426 water depth direct filter out the effect of changes in bottom albedo (e.g., Dierssen et al., measurements taken using a graduate tape attached on the bar holding 2003). Legleiter et al. (2004) and Marcus and Fonstad (2008) demon- the dGPS. strated that log-transformation of red-over-green band ratio linearly Corresponding colour band intensities and Zw were added to the correlates with water depth across a wide range of substrate types: points acquired in the wet areas (dGPS wet-area survey) obtaining a ¼ α þ β ðÞ= ðÞ shape file of points containing five fields (in addition to the spatial coor- DPH 0ln R G 2 dinates x and y): intensity of the three colour bands, Red (R), Green (G),

Blue (B), elevation of the channel bed (Zwet), and Zw. Finally, channel where DPH is the water depth [m], α and βx are the calibration coeffi- depth was calculated as Dph = Zw − Zwet [ma.s.l.]. These estimated cient, and R and G are the intensities of the red and green bands. water depths were validated by comparison to the water depth mea- An empirical linear model evaluating all the colour bands, possible sured in the field. A similar method was employed by Legleiter (2013) interactions and square, and cubic terms, were then tested: using the difference between mean water surface elevation and bed el- ¼ α þ β þ β þ β þ β þ β þ β þ β evation, both derived from a dGPS survey. DPH 0R 1G 2B 3RB 4RG 5GB 6RGB þ β 2 þ β 2 þ β 2 þ β 3 þ β 3 þ β 3 ð Þ 7R 8G 9B 10R 11G 12B 3 3.3. Determination of the best bathymetric model

where α and βx are the calibration coefficients in the depth-colour re- Starting from the obtained dataset, water depth (estimated indirect- gression. In this model, the significance of each component was tested ly) was considered as dependent variable, with the three intensity col- and deleted when the adopted statistical test (explained below) result- our bands (R, G and B) being independent variables. 80% of the ed negative. dataset was used for calibrating the depth-colour model (calibration The statistical regressions were performed in R® environment using points) and the remaining 20% to verify the efficiency and choose the two methods: the “traditional regression method” based on statistical J. Moretto et al. / Catena 122 (2014) 180–195 185 significance testing of each variable (P-value b 0.05), and the AICc index with an unreal slope variation (derived by curvature calculation) (Burnham and Anderson, 2002). outside the proposed curvature range were removed. In addition, non-surface points (outliers; b5 % of total points distribution) were 3.4. Hybrid DTM creation and validation also deleted. After the filtering of points, the “water depth model” (DPH — water The final HDTM represents a full integration between filtered LiDAR depth model) was finally obtained. For each point, the corresponding ground points and colour bathymetric points. LiDAR points were used in Zw [ma.s.l.] was subtracted to acquire the estimated river bed elevation all dry areas and up to 0.20 m of water depth, whereas the bathymetric (Zwet = Zw − DPH = [ma.s.l]). Hybrid DTMs (HDTM) were built up points were used in the remaining wet areas. The capacity of LiDAR sig- with a natural neighbour interpolator, integrating Zdry points (from nal to perform reliable ground points where water depth is lower than LiDAR) in the dry areas and in the first wet layer (0–0.20 m) and Zwet 0.20 m has been previously verified (Moretto et al., 2013a). For instance, points (from colour bathymetry) in the remaining wet areas. Charlton et al. (2003) and Cavalli and Tarolli (2011) have reported that Finally, the HDTM models were validated by using dGPS cross- LiDAR pulse is able to penetrate shallow water when the laser signal is sectional surveys. The error of each “control point” was derived consid- not distorted. The unreliability of colour bathymetry in representing ering the difference between elevation of the HDTM and corresponding water depths lower than 0.20 m when there is a strong colour variation elevation of the dGPS control points. at the cannel bottom (periphyton, exposed pebbles, woody debris, etc.), The accuracy of HDTMs was estimated separately for wet and dry has been verified in the field. In these conditions, remarkable errors may areas, also taking into account the dGPS error (available from the instru- be introduced (N ± 0.20 m) in depth estimations (Legleiter et al., 2009). ment for each point). The average uncertainty for both wet and dry The best bathymetric model for each year was therefore applied to areas was calculated averaging the error derived by each dGPS control the wet areas starting from 0.20 m of water depth down (measured point available for the correspondent dry or wet area. The total average from the water surface elevation raster) on the georeferenced images, uncertainty was calculated by weighting dry and wet uncertainties with to determine the “Raw Channel Depth Raster” (RDPH). The RDPH was the correspondent surfaces. The error for each water layer (every 0.20 m then transformed into points (4 points/m2) and filtered, as explained of depth) was also calculated. below, in order to delete uncorrected points, mainly due to sunlight re- flection, turbulence, strong periphyton presence, and elements (wood 3.5. Analysis of morphological changes or sediment) above the water surface. According to this approach, the filtered depth (DPH) model was finally obtained. The high-resolution HDTMs allowed the exploration of the morpho- To filter out possible incorrect points, a method based on the analysis logical effects of the flood events occurred between our considered of slope changes in neighbouring cells was adopted. Changes of local surveys. slope calculated among neighbouring cells were analysed through a The Geomorphic Change Detection 5.0 (GCD) software developed by semi-automatic method which uses a “curvature raster” (Curvature Wheaton et al., 2010 (http://gcd.joewheaton.org) was used to perform tool - ESRI® ArcMap 10 - Moore et al., 1991), obtaining a value of curva- reliable DEMs of Difference (DoDs). Elevation uncertainty associated ture (slope derivative) for each cell. The curvature tool calculates for with the DoDs was calculated in Matlab environment (Fuzzy Logic ap- each cell the second derivate value of the input surface (RDPH) on a plication) using an “ad hoc” FIS file and considering slope, point density cell-by-cell basis (3 × 3 moving window). For ranges of curvature and bathymetric points quality as input variables. Slope and point den- N700 or b−600, cells were considered incorrect outliers and conse- sity categorical limits (low, medium, high) were chosen taking into ac- quently eliminated. This range was derived from dGPS survey analy- count values available in the literature (Wheaton et al., 2010) and local sis and joined with direct observations over the estimated wet raster, environment. Bathymetric points quality was used to delete erroneous on which no changes in elevation greater than ± 0.60 m were pres- cells from the HDTMs in the final erosion–deposition volume computa- ent within a horizontal distance of b0.50 m. In other words, all areas tion (see Delai et al., 2014 for further details).

Table 1 Depth-colour model estimated by traditional and AICc method.

Model Depth-colour model estimated by traditional method p-value r2 Error (m)

Beer Lamb. DPH = −0.119 + 2.725 ln (R/G) 2.2 x 10-16 0.34 ±0.27 2010 Beer Lamb. DPH = −0.73 + 2.043 ln (R/G) 2.2 x 10-16 0.25 ±0.20 2011 Empirical DPH = 5.31 + 0.07513 R − 0.1869 G − 0.01475 B − 0.0004582 RB + 0.001056 G2 + 2.2 x 10-16 0.46 ±0.26 2010 0.0003352 B2 − 0.000002142 G3 Empirical DPH = −0.607 + 0.03508 R − 0.06376 G − 0.1377 B + 0.002257 RG − 0.001096 RB + 2.2 x 10-16 0.38 ±0.19 2011 0.002303 GB − 0.0007273 R2 − 0.002956 G2 +0.0009993 B2 +0.000002837 G3 − 0.00000685 B3

Model Depth-colour model estimated by AICc method p-value r2 Error (m)

Beer Lamb. DPH = −0.119 + 2.725 ln (R/G) n. a. n. a.±0.27 2010 Beer Lamb. DPH = −0.73 + 2.043 ln (R/G) n. a. n. a.±0.20 2011 Empirical DPH = 5.28 + 0.0000003527 R + 0.001189 G − 0.02082 B + 265 BR − 122.316 BG + n. a. n. a.±0.26 2010 0.073 RGB − 0.0008215 R2 − 0.000002506 G2 +0.0005809 B2 +10R3 +0.09595 G3 − 0.2026 B3 Empirical DPH = −0.607 + 0.03508 R − 0.06376 G − 0.1377 B + 0.002257 RG − 0.001096 RB + n. a. n. a.±0.19 2011 0.002303 GB − 0.0007273 R2 − 0.002956 G2 +0.0009993 B2 +0.000002837 G3 − 0.00000685 B3

Where DPH is the water depth [m], ln (R/G) are the colour bands arranged according to the Beer Lambert law. R, G and B are respectively the red, green and blue colour bands of the aerial photos. “P-value” and “square r” are parameters not available (n.a.) for AICc approach that has a complete different method of regression (Burnham and Anderson, 2002). The error (m) is derived from the model application on the 20 % of test points independents from the model calibration process with both statistical methods presented. 186 J. Moretto et al. / Catena 122 (2014) 180–195

Geomorphic changes in the study reaches were finally calculated, Canopy surface models (CSM), derived from the difference between similarly to Wheaton et al. (2010), by using a spatially variable uncer- digital surface models (DSM) and DTMs, were produced to identify nat- tainty thresholded at 95% C.I. and the Bayesian updating method ural (fluvial islands) and artificial (embankments and bridges) vertical which accounts for spatial coherent erosion and deposition units (5 × construction in the analysed sub-reaches. In addition to the bathymetric 5mobilewindows). rasters, three water depth classes (0–0.50 m; 0.50–1 m and N 1m)were A HDTMs comparison aimed at analysing the dynamics of the bed applied to identify different bed forms. forms (riffle–pool) as a consequence of flood events and natural and ar- tificial “constrictions”, was subsequently performed in order to inte- 4. Results grate erosion–deposition patterns analysis. 4.1. Colour bathymetry models

To understand the average and the range of channel depths, the av- erage, standard deviation and maximum depth of 2010 and 2011 wet channels were estimated before calibrating the regression model. 2010 was characterised by an average depth of 0.53 m, a standard devi- ation of 0.34 m and a maximum known depth of 1.62 m. 2011 had a greater average depth than 2010 equal to 0.63 m, a standard deviation of 0.28 m and a maximum known depth of 2.30 m. The indirect water depth estimation was validated thanks to 1426 di- rect measurements of water depth taken using a graduate tape attached on the bar holding the dGPS, and an average error of ±0.15 m was recognised. This error may be due to LiDAR vertical error (used to inter- polate the water surface), water turbulence around the graduated bar,

and by natural roughness of the bed surface (mainly cobbled; D84 = 64–87 mm, Moretto et al., 2012b). The search for the best depth-colour model started from the com- posed datasets by testing a physical model, based on the Beer Lambert law (Eq. (2)) for each year (2010 and 2011) and with the two intro- duced statistical regression methods (traditional regression and AICc index; Section 3.3). The application of the traditional regression method and the AICc index produced the same depth colour model for 2010 (see Beer Lam- bert 2010 equations on Table 1). This model has a statistically significant p-value ≪0.05, and an average error derived from the test points of ±0.27 m. A similar result was obtained for the 2011 model; also in this case the two statis- tical regression methods have produced the same result (see Beer Lam- bert 2011 equations on Table 1). This model has a statistically significant p-value ≪0.05, and an average error derived from the test points of ±0.20 m. The depth-colour (RGB) statistical regressions performed with the empirical model and using the two different approaches allowed two bathymetric models to be obtained for each year (2010 and 2011 empir- ical models — Table 1). The average errors detected in the two models by comparing the test points are equal to ±0.26 m. Negligible differences (0.003 m of differ- ence of average error) between the considered models (Table 1 — Em- pirical 2010 equations), traditional and AICc methods, were estimated. Therefore, the model resulting from the traditional method (see empir- ical 2010 equation in Table 1) was preferred because of its simpler struc- ture with fewer factors if compared to the AICc model. In these models (Table 1), DPH is the estimated water depth [m] and R, G and B are the red, green and blue bands, respectively. If 2011 is considered, the two different methodologies (traditional and AICc index) of statistical regression, generated the same equations as showed in Table 1 (Empirical 2011 equations). The estimated depth average error of 2011 resulting from the test points, accounts for ± 0.19 m. Both physical and empirical models proved to be statistically signif- icant (p-value ≪0.05), but the empirical models seem to have more predictive capacity than the physical approaches (see r2 on table 1). In addition, all three colour bands significantly contribute to depth estima- tion, therefore the presence of interactions between colour bands (as re- ported in Fig. 5) should be taken into consideration. Fig. 6 shows one of the outputs deriving from the model application Fig. 5. Correlation between Red, Green and Blue colour bands. (Table 1 — Empirical 2011 equations) in Friola sub-reach. It appears that J. Moretto et al. / Catena 122 (2014) 180–195 187

Fig. 6. Model application (8) at Friola sub-reach (2011). The brown zones on the left side are due to the presence of periphyton at the channel bottom. depth variations are generally respected, and variations in colour tone A cross-sections comparison of 2011 raw HDTM and the HDTM de- due to the presence of periphyton in the channel bed do not seem to rived from the profilesofFriolawetareasisshowninFig. 7. The sections strongly influence the estimation of water depth. In this sub-reach, the highlight the goodness of the applied filters. Also, the principal sources maximum estimated water depth exceeds 2 m. of error such as water turbulence, light reflections, suspended load, strong periphyton and exposed sediments appears to have been filtered 4.2. HDTM production and validation by the curvature calculation. The percentage of filtered depth points in Nove, Friola and Fontaniva The presented filters, used to delete raw depth points not belonging on 2010 wet areas were 3.39 %, 4.88 %, and 0.37 %, respectively. Instead, to water surface (due to model application on altered pixel colour the percentage of filtered depth points in Nove, Friola and Fontaniva on value) were applied on RDPH models. 2011 wet areas were 4.32 %, 2.60 %, and 18.11 %, respectively. In

Fig. 7. Example of filtering process in a cross-section of Friola sub-reach (2011). 188 J. Moretto et al. / Catena 122 (2014) 180–195

three for 2011 (Nove, Friola and Fontaniva sub-reaches) were generated using a 0.50 × 0.50 m cell size. Fig. 8 shows the HDTM obtained for Friola 2011. It is worth noticing the good representation of bed-forms such as riffles and pools within the wet channels. Data validation (Table 2) was performed separately for both wet and dry areas, obtaining average uncertainty values (by field survey com- parison) for each HDTM including dGPS, LiDAR and DPH estimated er- rors. Average uncertainty associated to wet areas accounts from a minimum of ± 0.19 m (Friola in 2011) to a maximum of ± 0.26 m (Nove and Fontaniva in 2010 and 2011), whereas in dry areas the aver- age uncertainty ranges from a minimum of ± 0.14 m (Nove in 2010) to a maximum of ± 0.26 m (Fontaniva in 2010). The chosen colour bathy- metric models (empirical depth-RGB) generated similar error levels, on dry and wet areas, for both 2010 and 2011. Moreover, the average weighted uncertainty was calculated in the final HDTMs, rang- ing from ± 0.16 m, for Nove 2010–2011 and Friola 2011, to ± 0.26 m in Fontaniva reach in 2010. If the errors associated with the HDTMs on wet areas are taken into consideration, the reliability of wet areas estimates in the HDTMs can be appreciated (Fig. 9). The percentage of control points within ± 0.30 m of error is equal to 75 % and 84 % for 2010 and 2011, respectively. Fig. 9 shows that the higher errors correspond to the maximum water depth and are up to 1 m and 1.20 m for 2010 and 2011, respectively. The dis- tribution of average errors, standard deviations and their aerial extent on the entire wet areas among different water depths are showed in Table 3. Fig. 10 reports an example of a comparison of three cross-sections for 2011, obtained with three different types of data (dGPS survey, LiDAR, and HDTM). The reference section was surveyed using a dGPS and ground points feature an average error of about 0.025 m. On the right hand side of Fig. 10 (zoom to the wet areas), we can ap- preciate a comparison between dGPS and LiDAR profiles: the ability of LiDAR signal to penetrate wet areas up to 0.25–0.30 m is confirmed. On the other hand, the use of LiDAR-derived water depth in channel areas deeper than 0.25–0.30 m can lead to underestimation of water depth, and a consequent overestimation of calculated DoD volumes, as Fig. 8. Hybrid Digital Terrain Model (HDTM) of Friola sub-reach, 2011, cell size 0.50 × 0.50 m. showed in Table 1. These volumes were derived as a subtraction be- tween HDTMs and DTMs (derived entirely from LiDAR). The minimum Fontaniva, more than 90,000 of the 499,596 depth raw points had to be volume of 397,470 m3 is registered at Friola, whereas the maximum of filtered out. This was the case in which more points needed filtering due 4,743,783 m3 at Fontaniva. Instead, comparing dGPS and HDTM profiles, to the presence of marked shadows generated by dense riparian vegeta- it appears that, overall, ground points are well replicated except for tion growing on banks (see Fig. 1). some small areas lower than the dGPS profiles (Fig. 10). This may be After filtering raw depth points, dry areas and the first 0.20 m of due, in part, to the presence of large boulders in the water channel water depths were integrated using the LiDAR flight. LiDAR point clouds that have altered the resulting cross-sections between precise dGPS (excluding wet areas) featured an average density of 2.07 points/m2 for measurements and those derived from a mediated profile by HDTM 2010 and 2.64 points/m2 for 2011. The final HDTMs, three for 2010 and cells of 0.50 × 0.50 m. The maximum registered depth is well replicated

Table 2 Estimated uncertainty for HDTM and DoD models.

NOVE FRIOLA FONTANIVA

2010 2011 2010 2011 2010 2011

HDTM area (m2) 566,916 566,916 836,967 836,967 627,049 627,049 Wet area (m2) 95,607 89,613 107,758 135,227 75,616 113,974 Wet area/HDTM area 0.17 0.16 0.13 0.16 0.12 0.18

N° dGPS point for test DTMBTH 192 408 279 821 204 283

Average unc. DTMBTH & dGPS (m) 0.26 0.26 0.25 0.19 0.26 0.26

N° dGPS point for test DTMLD 72 132 98 155 53 64

Average unc. DTMLD & dGPS (m) 0.14 0.15 0.24 0.15 0.26 0.16 TOTAL average uncertainty (m) 0.16 0.17 0.24 0.16 0.26 0.18 3 DoD (HDTM Vs. DTMFull LiDAR)(m) 917,559 529,812 1,206,848 397,470 4,386,814 4,743,783

DoD (HDTM2011 Vs. HDTM2010) Er. Dep. Er. Dep. Er. Dep. (m3) 122,498 18,416 177,951 95,030 158,359 113,127

DTMBTH: Part of Digital Elevation Model derived by Bathymetry; DTMLD: Part of Digital Elevation Model derived by Light Detection and Ranging; dGPS: Differential Global Positioning

System; DTMBTH or LD & dGPS: sum of DTM and dGPS error; DTMFull LiDAR: DTM totally derived from LiDAR; Er.: Erosion; Dep.: Deposition. J. Moretto et al. / Catena 122 (2014) 180–195 189

Fig. 9. Observed depth versus predicted depth classified by three level of errors:b±0.20 m;±0.20–0.30 m and N ±0.30 m. by colour bathymetry (comparing dGPS cross sections) and reaches 2 m are identified as dark areas, i.e. the zones with the higher water depth if as showed in Fig. 10c. compared to the riffles. It is worth noticing how the main channel ap- pears to have increased its sinuosity on the reaches where fewer lateral 4.3. Morphological change detection constrictions are present. Nove sub-reach is the most laterally constrained due to artificial left embankments featuring also the highest The effects of November and December 2010 floods on Nove sub- incision degree, and the main channel appears to have maintained the reach can be appreciated in Fig. 11a. Overall, erosion exceeds sedimen- same sinuosity. tation (122,498 vs. 18,416 m3; Table 2), and erosion consistently occurs Considering bed-forms after floods, pools appear to have increased along the main channel in the whole reach with a thickness of N0.20 m. in length. This is particularly evident in Friola sub-reach (pool P3 and Considering Friola sub-reach, the volumes of erosion and deposition P4, 2011) and Fontaniva (pool P4, 2011). The embankments and fluvial appear to be more similar than Nove, ranging around 177,951 m3 and islands appear to have played an important role in bed-form dynamics 95,030 m3, respectively (Table 2). Similarly, erosion occurs consistently during floods. Indeed, pools in each 2011 sub-reach are located mainly along the main channels with a thickness of N0.20–0.50 m. In Friola, ero- at the side of the wet area with a more compact lateral surface with em- sion of bars leading to lateral migration of the main channel can be bankments and/or vegetated bars. On the other hand, riffles are mainly depicted (Fig. 11b). located where no significant “constrictions” were present on either side In the lowermost sub-reach (Fontaniva) deposition and erosion are of the wet areas. volumetrically comparable, being around 158,359 m3 and 113,127 m3, respectively (Table 2). The erosion process is not continuous along the main channels as for Nove and Friola, but shows a more complex pat- 5. Discussion tern. Two different portions of the reach can be identified, being the upper one dominated by deposition, whereas the lower by predominant 5.1. Analysis of the proposed method for geomorphic change detection erosion. Fig. 12 shows the CSMs with the location of pools (e.g. P1, P2) on wet The proposed method is a revised procedure for the production of areas of Nove, Friola and Fontaniva sub-reaches in 2010 and 2011. Pools high resolution DTMs on gravel-bed rivers, integrating LiDAR points

Table 3 Error analysis of depth-colour models at different water stages for 2010 and 2011. The average error and standard deviation have been weighted with the correspondence inference area.

Depth Surface covered DPH (R, G, B) Survey method (m) (ha) % error (m) St. dev. (m)

2010) 0.00–0.19 9.23 33.1 0.26 0.22 LiDAR 0.20–0.39 6.88 27.7 0.26 0.24 Colour bathymetry 0.40–0.59 6.29 22.6 0.21 0.20 0.60–0.79 4.26 15.3 0.22 0.18 0.80–0.99 0.98 3.5 0.26 0.15 1.00–1.19 0.29 0.8 0.51 0.21 N1.20 0.03 0.1 0.69 0.14 TOTAL 27.90 100 0.25 0.21

2011) 0.00–0.19 0.17 0.5 0.27 0.11 LiDAR 0.20–0.39 2.32 6.9 0.18 0.11 Colour bathymetry 0.40–0.59 8.46 25.0 0.13 0.11 0.60–0.79 9.40 27.7 0.14 0.13 0.80–0.99 6.77 20.0 0.24 0.19 1.00–1.19 3.14 9.3 0.32 0.19 1.20–1.39 1.90 5.6 0.40 0.13 N1.40 1.72 5.1 0.56 0.10 TOTAL 33.89 100 0.21 0.14 190 J. Moretto et al. / Catena 122 (2014) 180–195

Fig. 10. Cross-section comparison on the left hand side between dGPS, HDTM and LiDAR surveys of Nove (a), Friola (b) and Fontaniva (c) 2011. Cross-section zoom in wet area on the right hand side. with filtered bathymetric points estimated through a regression model need direct field surveys of water depth because this is indirectly esti- implemented on wet areas. mated. Depth estimation entailed the subtraction of water level raster The bathymetric points can be derived from a physical and empirical (water surface) from corresponding dGPS elevation points (bottom sur- relationship between water depth and RGB bands of aerial images taken face) of the channel bed (Zwet). This method is an effective approach for concurrently with LiDAR data. indirect estimation of water depth and similar to the technique used by The model calibration requires a dGPS survey of water level, without Carbonneau et al. (2006). direct water depth measurements. It is crucial to acquire dGPS points Indirectly estimated depths (see Section 3.2), together with corre- nearly contemporary to LiDAR and aerial images, as already pointed sponding RGB values, are the values needed for the statistical calibration out by Legleiter (2011). In fact, the calibration of the model does not of the regression models. The statistical analysis showed that all

Fig. 11. Difference of DEMs (DoD) of Nove, Friola and Fontaniva sub-reaches. J. Moretto et al. / Catena 122 (2014) 180–195 191

Fig. 12. Canopy surface models (CSM) with pools individuation (P1, P2, etc.) through bathymetric raster on wet area in Nove, Friola and Fontaniva sub-reaches 2010 and 2011. 192 J. Moretto et al. / Catena 122 (2014) 180–195 three bands (R, G, B) and also some of the other additional factors (in- depth (about ± 0.40–0.60 m of difference with respect to the adjacent teractions among bands and square and cubic terms) were significant pixels). The correction method which involves the use of a filter based (p-value ≪ 0.05) to predict water depth. This statistical significance on curvature and consequent removal of outliers (points with errors ex- was also confirmed by two different statistical regression methods (ver- ceeding 95 % of confidence interval), has provided to work well as ification of p-value and AICc index). The “ad hoc” calibration for each depicted in Fig. 7. In this way the quality of the final HDTMs have study year was necessary because of the different water stage during been clearly improved. LiDAR survey. Shadows represent a disturbance factor difficult to correct and re- This study has demonstrated that in a wet area with complex and move because they tend to cause an overestimation of channel depth. heterogeneous channel bed, and with different colours on channel bot- However, their presence was minimal in the study sites, thanks to tom (due to the presence of periphyton), the tested depth-colour phys- image acquisition carried out around midday. The model tends to un- ical models does not perform as well as the empirical models. Indeed, derestimate water depth where this exceeds 1–1.10 m. This is partially the presence of periphyton on the channel bed can be claimed as one due to the low availability of calibration points (for safety reasons) in of the error sources (i.e. low r2) in assessing water depth, as previously the deepest areas of the water channel. Furthermore, in deeper water, suggested by similar studies (e.g. Carbonneau et al., 2006; Legleiter depth estimates through aerial images become less reliable due to the et al., 2011; Williams et al., 2014). Despite a lower r2,thefinal validation increase in saturation of the radiance signal (Legleiter, 2013). of the elevation models (shown in Fig. 10 and Table 2) has demonstrat- The main topographical variations, as showed in the comparison be- ed a bathymetric uncertainty comparable to LiDAR data in dry area. tween HDTMs and dGPS cross-sections (Fig. 10), result as being faithful- Table 3 shows that the optimal application range of the estimated ly reproduced, except for the thalweg, which was difficult to detect with bathymetric models is between 0.20 m and 1–1.20 m for 2010 and a dGPS survey. Consequently, the resulting HDTMs can be considered as 2011, respectively. Even though the best application range of estimated a satisfactory topographical representation (considering the resolution depths is lower than 1.20 m, local good estimations of wet channel up to of the final elevation models) for the homogeneous study of morpho- 2 m of water depth were observed in cross section comparisons logical variations. (Fig. 10c). The ability of LiDAR signal to allow a reliable estimate in the first 20 cm of water column was confirmed by dGPS and LiDAR cross- 5.2. Geomorphic changes after 2010 floods section comparisons (Fig. 10)aswellasinMoretto et al. (2013a). Smith et al. (2012) and Smith and Vericat (2013) report that a further The morphological evolution of the Brenta River over the last source of error could be due to laser refraction of LiDAR signal when 30 years has been strongly influenced by human impacts and flood penetrating into water. Although this error cannot be excluded in this events (Moretto et al., 2013b). Lateral annual adjustment is directly cor- study, it can be considered lower than other sources, because airborne related with the mean annual peak discharge (Moretto et al., 2012a, LiDAR signal has a near-vertical angle of incidence, thus presenting a 2013b), thus a higher flood magnitude generally corresponds to greater minimum level of refraction. Different errors at the same depth active channel widening. Substantial increases in channel width and re- (Table 3) between the two depth-colour models (one for each year) ductions of riparian vegetation occur due to flood events with an RI are likely due to the different number of calibration points among differ- higher than 5 years, as already highlighted by other studies on fluvial ent water depth ranges and different image luminosity conditions. In environments similar to the Brenta River (e.g. Bertoldi et al., 2009; Moretto et al. (2013a), the average error every 0.20 m of water depth as- Comiti et al., 2011; Kaless et al., 2014; Picco et al., 2012a,b, 2014). The sociated with the number of calibration points was calculated. A similar flood events of November–December 2010 in the Brenta River (RI = bathymetric approach was applied, but lower errors (b±0.26 m) were 8–10 years) have caused an increase of the active channel average assessed up to 1.60 m likely due to a larger amount of calibration points width of about 10 % (from 196 m to 215 m) with a consequent removal available. In this work, a calibration dataset as wide as the one used for of 10 ha of riparian vegetation in the study reach (for more detailed in- the Tagliamento River in Moretto et al. (2013a) was unfortunately not formation see Moretto et al., 2012a,b). It is worth pointing out that the available. Therefore, a preliminary analysis aimed at assessing both per- predominance of erosion processes, with a consequence negative bal- centage of present wet surface in the study area and range of depths, is ance between erosion and deposition, decreases from the upper to the required. In this way, a minimum number of calibration points for each lower sub-reach and is equal to −104,082 m3, −82,921 m3 and water depth range can be decided, allowing an acceptable error for most −45,232 m3, respectively. part of the wet areas. It is interesting to note the presence of a continuous eroded layer Despite the possible sources of errors, the proposed approach along the main channels (0.20–0.50 m of depth) along Nove and Friola allowed to generate elevation models with a vertical error lower than sub-reaches. Instead, in Fontaniva sub-reach the upper part features ± 0.22 m for 95.6 % of the 2010 wet area and lower than ± 0.26 m for predominant deposition dynamics, whereas erosion seems to dominate 99.1 % of the same surface. For 2011, we obtained vertical errors lower in the lower part. than ± 0.24 m for 80.0 % and lower than ± 0.32 m for 89.3 % of the The analysed flood events seem to have generated riffle–pool migra- wet area, respectively. Hydraulic conditions at the time of LiDAR survey tions on unconfined sections (e.g. P1 on Friola and Fontaniva 2010), were not exactly the same in 2010 and 2011 (see Section 4.1), and the while a pool enlargement occurred beside an artificial lateral constric- number of calibration points can play a significant role especially in a tion (e.g. P4 on Nove and P3–P4 on Friola 2011). The location and geom- very variable fluvial environment. etry of the new bed forms seem to be related to the natural (vegetated The importance of using a bathymetric method for evaluating ero- bar) and anthropic (embankments and bridges) constrictions. If pools sion–deposition patterns by applying numerical models or developing are compared from 2010 to 2011, it appears that after a severe flood sediment budgets is showed in Table 2 where the loss of volume with- event, they are generally longer and migrations are more concentrated out applying colour bathymetry is reported. on reaches partially or totally confined (Fig. 12). In Fig. 7, different types of errors were identified in the raw HDTM: The different behaviour of the three sub-reaches seems to be attrib- light reflection, water turbulence, periphyton and exposed sediment utable to their different morphological characteristics (natural and im- (sources of errors highlighted also by Legleiter et al., 2009). Light reflec- posed) and the availability of sediment from the upstream reach tion and water turbulence (white pixels) produce strongly negative (Moretto et al., 2012a,b, 2013b). The first sub-reach (Nove) is the depth estimates and substantially different (about 1–2 m) values from most affected by erosion processes (Moretto et al., 2012a). The condi- adjacent pixels not affected by these problems. Exposed or nearly ex- tions of Nove sub-reach can be summarised as follows: i) past and pres- posed periphyton (green and brown pixels) and exposed sediment ent heavy incision of the active channel with modifications in section (grey pixels) produce an underestimation or overestimation of water shape and from the river basin; ii) very little sediment supply from J. Moretto et al. / Catena 122 (2014) 180–195 193 upstream reaches; iii) almost total absence of vegetation on the flood- The validation of the Hybrid Digital Terrain Models (HDTM) resulted plain; iv) increase of local slope. satisfactory for distributed evaluations of morphological variations. The The second sub-reach, Friola, has a lower slope and is less laterally bathymetric method proved to be fundamental to obtain realistic eval- constrained than the upstream area, as confirmed by the presence of a uations when aiming at quantifying erosion–deposition patterns, apply- large island and a secondary channel on the right side. During severe ing numerical models to simulated floods or developing sediment floods, therefore, the main channel can migrate forming new deposition budgets. bars. On the other hand, the dynamics of Fontaniva are related to: i) The flood events of November–December 2010 (RI = 8 and greater availability of eroded sediment coming from the upper sub- 10 years) have caused significant geomorphic changes in all three reaches; ii) more balanced erosion-deposition pattern (Moretto et al., sub-reaches. The different behaviour among the sub-reaches seems to 2012a,b, 2013b); iii) increase in the average elevation of the active be attributable to their diverse morphological characteristics and the channel in the last 30 years; iv) presence of extended and stable vege- availability of sediment from upstream. A predominance of erosion pro- tation in the floodplain area which is increasingly affected by flood cesses, with a consequent negative balance between deposition and events; v) reduction of local slope; vi) presence of infrastructures (2 erosion at sub-reach level decreasing when going from the upper bridges). The reduction of slope, together with the vertical aggradation reach (−104,082 m3) to the lower one (−45,232 m3), was found. Rif- of the active channel over the last 30 years (Moretto et al., 2012a,b, fle–pool dynamics seem influenced by the nature of lateral constriction 2013b), determine a greater lateral mobility due to flood events, espe- (natural banks vs. embankments and bridges). After a severe flood cially on the banks with dense and stable riparian vegetation (Fig. 1). event, pools seem be located mainly near compact banks with embank- The morphological changes that occurred in the Brenta River as a ments and/or vegetated bars. On the other hand, riffles seem to be locat- consequence of the flood events in 2010 (RI of about 8 and 10 years) ed mainly where no significant constrictions were present on either side are of interest to evaluate the fluvial hydro-morphological quality, be- of the wet areas. cause they highlight the processes that are taking place, and provide in- The results of this study can be a valuable support to generate pre- sights for their future evolution as required by the EU Water Framework cise elevation models also for wet areas, useful for evaluating erosion– Directive. Nevertheless, the implementation of evolutionary models and deposition patterns, improving sediment budget calculations and the the estimation of sediment transport require a better assessment of the implementation of 2D and 3D numerical hydrodynamic models. quantity of incoming and outbound sediment in the study reach and a detailed analysis of the transport rate in relation to the event magni- Notation tude. Several studies tried to apply the morphological approach for esti- mating sediment budget starting from transversal sections (i.e. Bertoldi et al., 2010; Lane, 1998; Surian and Cisotto, 2007), even if a much more accurate spatial definition can be obtained from remote sensing data dGPS Differential Global Positioning System (i.e. Hicks, 2012; Hicks et al., 2006; Milan and Heritage, 2012; Rennie, DEM Digital Elevation Model 2012). The traditional methodologies of terrain change detection (e.g. DPH Channel Depth [m] with dGPS cross-sections) provide higher local precision, but the deter- DTM Digital Terrain Model mination of volume changes at reach scale may be improved with the HDTM Hybrid Digital Terrain Model assessment of DEMs differences (Lane et al., 2003). The implementation LiDAR Light Detection And Ranging of LiDAR data and colour bathymetry with the proposed methodology RDPH Raw channel depth model (raster and/or points) allowed us to obtain a terrain digital model with sufficient accuracy to RGB Red Green Blue derive patterns of sediment transfer, in particular within the water RI Recurrence Interval [years] channels. The information obtained from such analyses should be inte- Zdry Z coordinate of dry area [m.a.s.l.] grated with direct field measurements. Zwet Z coordinate of wet area [m.a.s.l.] Zw Z coordinate of water level [m.a.s.l.] 6. Final remarks

The proposed methodology allows the production of high-resolution Acknowledgements DTMs of wet areas with an associated uncertainty that has proved to be comparable to the LiDAR data up to 1–1.20 m of water depth. The bathy- This research was funded by the CARIPARO Research Project metric model calibration requires only a dGPS survey in the wet areas “Linking geomorphological processes and vegetation dynamics in taken during aerial image acquisition. Statistical analyses have demon- gravel-bed rivers”; the University of Padua Strategic Research Project strated that all three colour bands (R, G, B) significantly correlate with PRST08001, “GEORISKS, Geological, morphological and hydrological water depth with a good performance of the empirical models. In addi- processes: monitoring, modelling and impact in North-Eastern Italy”, tion, the presence of an interaction between the colour bands cannot be Research Unit STPD08RWBY-004; the Italian National Research Project neglected. This study supports the evidence that, in a complex gravel- PRIN20104ALME4-ITSedErosion: “National network for monitoring, bed river with different water depths and different colours on channel modelling and sustainable management of erosion processes in agricul- bottom, the tested physical models have a lower degree of significance tural land and hilly-mountainous area”; and the EU SedAlp Project: in respect to the empirical models. Different errors were identified on “Sediment management in Alpine basins: Integrating sediment contin- raw HDTM: light reflections, water turbulences, strong colour variations uum, risk mitigation and hydropower”, 83-4-3-AT, within the frame- at the bottom, periphyton, shadows, suspended load, exposed sediment work of the European Territorial Cooperation Programme Alpine and submerged vegetation. Space 2007–2013. All colleagues and students who helped in the field Error sources were mostly intercepted through two proposed filters are sincerely thanked. Thanks to mother tongue Ms Alison Garside for which consider curvature assessment and implausible upper and lower the efficient English correction of the final English version and for the limits in the bathymetric raster. As a consequence, a preliminary analy- final paper check. sis seems to be needed to assess in advance both the percentage of wet surface and the range of depths in the river reach to be surveyed in order to reduce the number of calibration points for each water depth References class, thus allowing a fairly acceptable error for the major part of the Ashmore, P.E., Church, M.J., 1998. Sediment transport and river morphology: a paradigm wet areas. for study. In: Kingleman, P.C., Bechta, R.L., Komar, P.D., Bradley, J.B. (Eds.), Gravel Bed 194 J. Moretto et al. / Catena 122 (2014) 180–195

Rivers in the Environment. Highland Ranch, CO, Water Resources Publications, pp. Kinzel, P.J., Wright, C.W., Nelson, J.M., Burman, A.R., 2007. Evaluation of an experimental 115–148. LiDAR for surveying a shallow, braided, sand-bedded river. J. Hydraul. Eng. 133, Bathurst, J.C., Crosta, G.B., Garcìa Ruiz, J.M., Guzzetti, F., Lenzi, M.A., Aragues, S., 2003. De- 838–842. bris fall assessment in mountain catchments for local End-users. International Con- Kinzel, P.J., Legleiter, C.J., Nelson, J.M., 2013. Mapping River Bathymetry With a Small Foot- ference on Debris Flow Hazards Mitigation: Mechanics, Prediction and Assessment, print Green LiDAR: applications and Challenges. J. Am. Water Resour. Assoc. 49, Proceedings 2, pp. 1073–1083. 183–204. http://dx.doi.org/10.1111/jawr.12008. Bertoldi, W., Gurnell, A., Surian, N., Tockner, K., Zanoni, L., Ziliani, L., Zolezzi, G., 2009. Un- Lane, S.N., 1998. The use of digital terrain modelling in the understanding of dynamic derstanding reference processes: linkages between river flows, sediment dynamics river channel systems. In: Lane, S.N., Richards, K., Chandler, J. (Eds.), Landform Mon- and vegetated landforms along the Tagliamento River, Italy. River Res. Appl. 25, itoring, Modelling and Analysis. Wiley, Chichester, pp. 311–342. 501–516. Lane, S.N., Richards, K.S., Chandler, J.H., 1994. Developments in monitoring and terrain Bertoldi, W., Zanoni, L., Tubino, M., 2010. Assessment of morphological changes induced modelling of small-scale riverbed topography. Earth Surf. Process. Landf. 19, by flow and flood pulses in a gravel bed braided river: the Tagliamento River 349 –368. (Italy). Geomorphology 114, 348–360. Lane, S.N., Westaway, R.M., Hicks, D.M., 2003. Estimation of erosion and deposition vol- Brasington, J., Rumsby, B.T., McVey, R.A., 2000. Monitoring and modelling morphological umes in a large, gravel-bed, braided river using synoptic remote sensing. Earth Surf. change in a braided gravel-bed river using high resolution GPS-based survey. Earth Process. Landf. 28 (3), 249–271. http://dx.doi.org/10.1002/esp.483. Surf. Process. Landf. 25 (9), 973–990. Lane, S.N., Tayefi, V., Reid, S.C., Yu, D., Hardy, R.J., 2007. Interactions between sediment de- Brasington, J., Langham, J., Rumsby, B., 2003. Methodological sensitivity of morpho- livery, channel change, climate change and flood risk in a temperate upland environ- metric estimates of coarse fluvial sediment transport. Geomorphology 53 (3), ment. Earth Surf. Process. Landf. 32, 429–446. 299–316. Lane, S.N., Widdison, P.E., Thomas, R.E., Ashworth, P.J., Best, J.L., Lunt, I.A., Sambrook Smith, Brasington, J., Vericat, D., Rychkov, I., 2012. Modeling river bed morphology, roughness, G.H., Simpson, C.J., 2010. Quantification of braided river channel change using archi- and surface sedimentology using high resolution terrestrial laser scanning. Water val digital image analysis. Earth Surf. Process. Landf. 35, 971–985. http://dx.doi.org/ Resour. Res. 48 (11). http://dx.doi.org/10.1029/2012WR012223. 10.1002/esp.2015. Buffington, J.M., 2012. Change in channel morphology over human time scales. In: Legleiter, C.J., 2011. Remote measurement of river morphology via fusion of LiDAR topog- Church, M., Biron, P.M., Roy, A.G. (Eds.), Gravel-bed Rivers: Processes, Tool, Environ- raphy and spectrally based bathymetry. Earth Surf. Process. Landf. 37, 499–518. ments. Wiley-Blackwell, pp. 435–463. Legleiter, C.J., 2013. Mapping river depth from publicly available aerial images. River Res. Burnham, K.P., Anderson, D.R., 2002. 2nd ed. Model Selection and Multimodel Inference: Appl. 29, 760–780. http://dx.doi.org/10.1002/rra.2560. A Practical Information-Theoretic Approach, 16. Springer, p. 488. Legleiter, C.J., Roberts, D.A., 2009. A forward image model for passive optical remote sens- Carbonneau, P.E., Lane, S.N., Bergeron, N.E., 2006. Feature based image processing ing of river bathymetry. Remote Sens. Environ. 113, 1025–1045. methods applied to bathymetric measurements from airborne remote sensing in flu- Legleiter, C.J., Roberts, D.A., Marcus, W.A., Fonstad, M.A., 2004. Passive optical remote vial environments. Earth Surf. Process. Landf. 31, 1413–1423. http://dx.doi.org/10. sensing of river channel morphology and in-stream habitat: physical basis and feasi- 1002/esp.1341. bility. Remote Sens. Environ. 93, 493–510. Cavalli, M., Tarolli, P., 2011. Application of LiDAR technology for rivers analysis. Ital. J. Eng. Legleiter, C.J., Roberts, D.A., Lawrence, R.L., 2009. Spectrally based remote sensing of river Geol. Environ. 11, 33–44. bathymetry. Earth Surf. Process. Landf. 34, 1039–1059. Charlton, M.E., Large, A.R., Fuller, I.C., 2003. Application of airborne LiDAR in river environ- Legleiter, C.J., Kinzel, P.J., Overstreet, B.T., 2011. Evaluating the potential for remote bathy- ments: the River Coquet, Northumberland, UK. Earth Surf. Process. Landf. 28 (3), metric mapping of a turbid, sand-bed river: 1. Field spectroscopy and radiative trans- 299–306. fer modeling. Water Resour. Res. 4, W09531. Comiti, F., 2011. How natural are Alpine mountain rivers? Evidence from the Italian Alps. Lenzi, M.A., 2006. Research developments in debris flow monitoring, modeling and hazard Earth Surf. Process. Landf. 37, 693–707. http://dx.doi.org/10.1002/esp.2267. assessment in Italian mountain catchments. WIT Trans. Ecol. Environ. 90, 135–145. Comiti, F., Da Canal, M., Surian, N., Mao, L., Picco, L., Lenzi, M.A., 2011. Channel adjust- Lenzi, M.A., D'Agostino, V., Gregoretti, C., Sonda, D., 2003. A simplified numerical model ments and vegetation cover dynamics in a large gravel bed river over the last for debris-flow hazard assessment: DEFLIMO. International Conference on Debris 200 years. Geomorphology 125, 147–159. Flow Hazards Mitigation: Mechanics, Prediction and Assessment, Proceedings 1, Conesa-Garcìa, C., Lenzi, M.A., 2013. Check Dams, Morphological Adjustments and Erosion pp. 611–622. Control in Torrential Streams. Nova Science Publishers, Inc., New York 978-1-60876- Lenzi, M.A., Mao, L., Comiti, F., Rigon, E., Picco, L., Vitti, P., Moretto, J., Sigolo, C., 2010. Sci- 146-3, pp. 1–339. entific contribution by the Research Unit Land and Agro-forest Department, to the re- Delai, F., Moretto, J., Picco, L., Rigon, E., Ravazzolo, D., Lenzi, M.A., 2014. Analysis of search activities carried out in the framework of the CARIPARO Project “Linking morphological processes in a disturbed gravel-bed river (Piave River): integration geomorphological processes and vegetation dynamics in gravel-bed rivers”, from of LiDAR data and colour bathymetry. J. Civil Eng. Archit. 8 (5), 639–648 (ISSN September 2009 to October 2010. Research and Technical Report. Department of 1934-7359). Land and Agro-forest Environment, University of Padova, Padova, Italy, p. 102. Dierssen, H.M., Zimmerman, R.C., Leathers, R.A., Downes, T.V., Davis, C.O., 2003. Macklin, M.G., Rumsby, B.T., 2007. Changing climate and extreme floods in the British up- Ocean color remote sensing of seagrass and bathymetry in the Bahamas lands. Trans. Inst. Br. Geogr. 32, 168–186. http://dx.doi.org/10.1111/j.1475-5661. Banks by high-resolution airborne imagery. Limnol. Oceanogr. 48 (1, part 2), 2007.00248.x. 444–455. Mao, L., Lenzi, M.A., 2007. Sediment mobility and bedload transport conditions in an al- Dixon, L.F.J., Barker, R., Bray, M., Farres, P., Hooke, J., Inkpen, R., Merel, A., Payne, D., pine stream. Hydrol. Process. 21 (14), 1882–1891. Shelford, A., 1998. Analytical photogrammetry for geomorphological research. In: Mao, L., Cavalli, M., Comiti, F., Marchi, L., Lenzi, M.A., Arattano, M., 2009. Sediment transfer Lane, S.N., Richards, K.S., Chandler, J.H. (Eds.), Landform Monitoring, Modelling and processes in two Alpine catchments of contrasting morphology settings. J. Hydrol. Analysis. John Wiley & Sons, Chichester, pp. 63–94 (Chapter 4). 364, 88–98. Fewtrell, T.J., Duncan, A., Sampson, C.C., Neal, J.C., Bates, P.D., 2011. Benchmarking urban Marcus, W.A., 2012. Remote sensing of the hydraulic environments in gravel-bed rivers. flood models of varying complexity and scale using high resolution terrestrial In: Church, M., Biron, P.M., Roy, A.G. (Eds.), Gravel-bed Rivers: Processes, Tool, Envi- LiDAR data. Phys. Chem. Earth A/B/C 36 (7), 281–291. ronments. Wiley-Blackwell, pp. 261–285. Fryer, J.G., 1983. A simple system for photogrammetric mapping in shallow-water. Marcus, W.A., Fonstad, M.A., 2008. Optical remote mapping of rivers at sub-meter resolu- Photogramm. Rec. 11, 203–208. tions and watershed extents. Earth Surf. Process. Landf. 33, 4–24. Heritage, G.L., Fuller, I.C., Charlton, M.E., Brewer, P.A., Passmore, D.P., 1998. CDW photo- Marcus, W.A., Legleiter, C.J., Aspinall, R.J., Boardman, J.W., Crabtree, R.L., 2003. High spatial grammetry of low relief fluvial features: accuracy and implications for reach-scale resolution hyperspectral mapping of in-stream habitats, depths, and woody debris in sediment budgeting. Earth Surf. Process. Landf. 23, 1219–1233. mountain streams. Geomorphology 55, 363–380. Hicks, D.M., 2012. Remotely sensed topographic change in gravel riverbeds with flowing McKean, J.A., Nagel, D., Tonina, D., Bailey, P., Wright, C.W., Bohn, C., Nayegandhi, A., 2009. channels. In: Church, M., Biron, P.M., Roy, A.G. (Eds.), Gravel-bed Rivers: Processes, Remote sensing of channels and riparian zones with a narrow-beam aquatic- Tool, Environments. Wiley-Blackwell, pp. 303–314. terrestrial LIDAR. Remote Sens. 1 (4), 1065–1096. Hicks, D.M., Duncan, M.J., Walsh, J.M., Westaway, R.M., Lane, S.N., 2002. New views of the Milan, D.J., Heritage, G.L., 2012. LiDAR and ADCP use in gravel-bed rivers: advances since morphodynamics of large braided rivers from high-resolution topographic surveys GBR6. In: Church, M., Biron, P.M., Roy, A.G. (Eds.), Gravel-bed Rivers: Processes, Tool, and time-lapse video. IAHS Publ. 276, 373–380. Environments. Wiley-Blackwell, pp. 286–302. Hicks, D.M., Shankar, U., Duncan, M.J., Rebuffe, M., Abele, J., 2006. Use of remote sensing Milan, D.J., Heritage, G.L., Hetherington, D., 2007. Application of a 3D laser scanner in the technologies to assess impacts of hydro-operations on a large, braided, gravel-bed assessment of erosion and deposition volumes and channel change in a proglacial river: Waitaki River, New Zealand. In: Sambrook Smith, G.H., Best, J.L., Bristow, C.S., river. Earth Surf. Process. Landf. 32, 1657–1674. Petts, G.E. (Eds.), Braided Rivers, Processes, Deposits, Ecology and Management. In- Milan, D.J., Heritage, G.L., Large, A.R.G., Fuller, I.C., 2011. Filtering spatial error from DEMs: ternational Association of Sedimentologists, Special Publication, 36. Blackwell, Ox- implications for morphological change estimation. Geomorphology 125, 160–171. ford, pp. 311–326. Moore, I.D., Grayson, R.B., Landson, A.R., 1991. Digital terrain modelling: a review of hy- Hilldale, R.C., Raff, D., 2008. Assessing the ability of airborne LiDAR to map river bathym- drological, geomorphological, and biological applications. Hydrol. Process. 5, 3–30. etry. Earth Surf. Process. Landf. 33, 773–783. Moretto, J., Rigon, E., Mao, L., Picco, L., Delai, F., Lenzi, M.A., 2012a. Medium-and short- Kaless, G., Mao, L., Lenzi, M.A., 2011. Regime theories in gravel bed rivers; preliminary term channel and island evolution in a disturbed gravel bed river (Brenta River, comparison between disturbed rivers due to antrophic activities (Northeastern Italy). J. Agric. Eng. 43 (4), 176–188. http://dx.doi.org/10.4081/jae.2012.e27. Italy) and natural rivers (Patagonia, Argentina). Proceedings of the Intermediate Con- Moretto, J., Rigon, E., Mao, L., Picco, L., Delai, F., Lenzi, M.A., 2012b. Assessing short term gress of the Italian Association of Agricultural Engineering; Belgirate, Italy; September erosion-deposition processes of the Brenta River using LiDAR survey. WIT Trans. 22–24, 2011, p. 8. Eng. Sci. 73, 149–160. http://dx.doi.org/10.2495/DEB120131. Kaless, G., Mao, L., Lenzi, M.A., 2014. Regime theories in gravel bed rivers: models, control- Moretto, J., Delai, F., Picco, L., Lenzi, M.A., 2013a. Integration of colour bathymetry, LiDAR ling variables, and applications in disturbed Italian rivers. Hydrol. Process. 28, and dGPS surveys for assessing fluvial changes after flood events in the Tagliamento 2348–2360. http://dx.doi.org/10.1002/hyp.9775. River (Italy). Agric. Sci. 4 (8A), 21–29. http://dx.doi.org/10.4236/as.2013.48A004. J. Moretto et al. / Catena 122 (2014) 180–195 195

Moretto, J., Rigon, E., Mao, L., Picco, L., Delai, F., Lenzi, M.A., 2013b. Channel adjustment Sampson, C.C., Fewtrell, T.J., Duncan, A., Shaad, K., Horritt, M.S., Bates, P.D., 2012. Use of and island dynamics in the Brenta River (Italy) over the last 30 years. River Res. terrestrial laser scanning data to drive decimetric resolution urban inundation Appl. http://dx.doi.org/10.1002/rra.2676. models. Adv. Water Resour. 41, 1–17. Muste, M., Kim, D., Merwade, V., 2012. Modern digital instruments and techniques for hy- Smith, M.W., Vericat, D., 2013. Evaluating shallow-water bathymetry from through- drodynamic and morphologic characterization of river channels. In: Church, M., water terrestrial laser scanning under a range of hydraulic and physical water Biron, P.M., Roy, A.G. (Eds.), Gravel-bed Rivers: Processes, Tool, Environments. quality conditions. River Res. Appl. http://dx.doi.org/10.1002/rra.2687 (pub- Wiley-Blackwell, pp. 315–341. lished online 10 July 2013). Panissod, F., Bailly, J.S., Durrieu, S., Jacome, A., Mathys, N., Cavalli, M., Puech, C., 2009. Smith, M., Vericat, D., Gibbins, C., 2012. Through water terrestrial laser scanning of gravel Qualifiction de modeles numeriques de terrain Lidar pour l'etude de l'erosion: Appli- beds at the patch scale. Earth Surf. Process. Landf. 37 (4), 411–421. cation aux badlands de draix. Rev. Fr. Photogramm. Teledetection 192, 50–57. Stover, S.C., Montgomery, D.R., 2001. Channel change and flooding, Skokomish River, Picco, L., Mao, L., Rigon, E., Moretto, J., Ravazzolo, D., Delai, F., Lenzi, M.A., 2012a. Medium Washington. J. Hydrol. 243, 272–286. term fluvial island evolution in relation with floods events in the Piave River. WIT Surian, N., Cisotto, A., 2007. Channel adjustments, bedload transport and sediment Trans. Eng. Sci. vol. 73, 161–172. http://dx.doi.org/10.2495/DEB120141 (ISSN sources in a gravel-bed river, Brenta river, Italy. Earth Surf. Process. Landf. 32, 1743–3522). 1641–1656. Picco, L., Mao, L., Rigon, E., Moretto, J., Ravazzolo, D., Delai, F., Lenzi, M.A., 2012b. Riparian Surian, N., Ziliani, L., Comiti, F., Lenzi, M.A., Mao, L., 2009. Channel adjustments and alter- forest structure, vegetation cover and flood events in the Piave River. WIT Trans. Eng. ation of sediment fluxes in gravel-bed rivers of north-eastern Italy: potentials and Sci. 73, 137–147. http://dx.doi.org/10.2495/DEB120121 (ISSN 1743–3522). limitations for channel recovery. River Res. Appl. 25, 551–567. Picco, L., Mao, L., Cavalli, M., Buzzi, R., Rainato, R., Lenzi, M.A., 2013. Evaluating short-term Taylor, J., 1997. An Introduction to Error Analysis: The Study of Uncertainties in Physical morphological changes in a gravel-bed braided river using Terrestrial Laser Scanner. Measurements, 2nd ed. University Science Books, Sausalito, CA. Geomorphology 201, 323–334. http://dx.doi.org/10.1016/j.geomorph.2013.07.007. Wheaton, J.M., Brasington, J., Darby, S.E., Sear, D.A., 2010. Accounting for uncertainty in Picco, L., Mao, L., Rainato, R., Lenzi, M.A., 2014. Medium-term fluvial island evolution in a DEMs from repeat topographic surveys: improved sediment budgets. Earth Surf. Pro- disturbed gravel-bed river (Piave River, Northeastern Italian Alps). Geogr. Ann. Ser. A cess. Landf. 35, 136–156. Phys.Geogr.96,83–97. http://dx.doi.org/10.1111/geoa.12034. Wheaton, J.M., Brasington, J., Darby, S.E., Kasprak, A., Sear, D., Vericat, D., 2013. Rennie, C.D., 2012. Mapping water and sediment flux distributions in gravel-bed rivers Morphodynamic signatures of braiding mechanisms as expressed through change using ADCPs. In: Church, M., Biron, P.M., Roy, A.G. (Eds.), Gravel-bed Rivers: Process- in sediment storage in a gravel‐bed river. J. Geophys. Res. Earth Surf. 118 (2), es, Tool, Environments. Wiley-Blackwell, pp. 342–350. 759–779. Rigon, E., Comiti, F., Mao, L., Lenzi, M.A., 2008. Relationships among basin area, sediment Williams, R.D., Brasington, J., Vericat, D., Hicks, D.M., Labrosse, F., Neal, M., 2011. Monitor- transport mechanisms and wood storage in mountain basins of the Dolomites (Italian ing braided river change using terrestrial laser scanning and optical bathymetric Alps). WIT Trans. Eng. Sci. 60, 163–172. mapping. In: Smith, M., Paron, P., Griffiths, J. (Eds.), Geomorphological Mapping. Rigon, E., Comiti, F., Lenzi, M.A., 2012. Large wood storage in streams of the Eastern Italian Elsevier, Oxford, pp. 507–531. Alps and the relevance of hillslope process. Water Resour. Res. 48, W01518. http://dx. Williams, R.D., Brasington, J., Hicks, M., Measures, R., Rennie, C.D., Vericat, D., 2013. Hy- doi.org/10.1025/2010 WR009854. draulic validation of two‐dimensional simulations of braided river flow with spatially Rinner, K., 1969. Problems of two medium photogrammetry. Photogramm. Eng. 35, 275. continuous aDcp data. Water Resour. Res. 49 (9), 5183–5205. Rumsby, B.T., Macklin, M.G., 1994. Channel and floodplain response to recent abrupt cli- Williams, R.D., Brasington, J., Vericat, D., Hicks, D.M., 2014. Hyperscale terrain modelling mate change: the Tyne basin, Northern England. Earth Surf. Process. Landf. 19 (6), of braided rivers: fusing mobile terrestrial laser scanning and optical bathymetric 499–515. mapping. Earth Surf. Process. Landf. 39 (2), 167–183. Rumsby, B.T., Brasington, J., Langham, J.A., McLelland, S.J., Middleton, R., Rollingson, G., Winterbottom, S.J., Gilvear, D.J., 1997. Quantification of channel bed morphology in 2008. Monitoring and modelling particle and reach-scale morphological change in gravel-bed rivers using airborne multispectral imagery and aerial photography. gravel-bed rivers: applications and challenges. Geomorphology 93 (1), 40–54. Regul. Rivers Res. Manag. 13, 489–499.